Study On 3D Reconstruction Of Medical X-CT Image

The appearance of one of the greatest achievement in medical research in 20th Century —— Computerized Tomography, have brought a revolutionary change to radiotherapy research. With the development of the modern medical imaging technology and the computer technology, and their interaction as well, to design or product a 3D visualized medical accessory system has become reality. At present, 3D reconstruction from medical images is an extensive-research subject with multi-disciplinary interaction, and it is an important application of computer graphics and image processing in biomedicine engineering. It relates to the subjects of digital image processing, computer graphics and some related knowledge of medical. 3D reconstruction and visualization from medical images are widely used in diagnostic, surgery planning and simulation, plastic and artificial limb surgery, radiotherapy planning, and teaching in anatomy.This dissertation summarizes the present status, both domestic and abroad, of medical image visualization. In this dissertation, the key technologies of 3D reconstruction from medical CT are studied, and some concerned principles and algorithms are analyzed, including image preprocessing, segmentation and extraction of human tissues or organs, 3D surfaces reconstruction from contours, 3D models construction by isosurface extracting, meshes simplifying of the reconstructed models, and so on. This paper analyzes the shortage of the normal MC algorithm that causes huge consumption of operation and is hard to work out, then presents SMC method which aims at the 3D reconstruction from medical spiral CT images. With SMC, a 3D reconstruction PC system from medical CT has been set up, this dissertation explicates the design and methods of the system.This dissertation has two features below:(1) According to the shortage of normal MC algorithm, a surface-fitting method modified from normal MC is presented. This method defines a status function with 3 status(+1,0,-1), with which it need only judging the status of each vertex in the chamber to reconstruct the surfaces by using the character that a pixel has the same sign as its 4 neighborhood in one CT image. This method avoids calculating the asymptote of hyperbola to judge the ambiguity, and reduces the complexity of reconstruction. (2) Using edge deleting algorithm to simplify the meshes of the reconstructed models, calculating the priority by the weighted mean of multi-element, and taking the flatness level of meshes as the chief factor. This will shorten the operating time of meshes simplifying.